bug: stonedb OOM use parallel hash join when the table is very large(100G)

[adofsauron]

Describe the problem

1. when force use parallel hash join and the table is very large beyond the memory
2. the stonedb on paralller hash join will OOM and kill by kernel

Expected behavior

1. if table size beyond memory, parallel will process fastly and safely together

How To Reproduce

1. at this time, could use serial hash join ignore this OOM, but I suggesst process parallel on the high efficiency

Environment

1. OS: centos7.5
2. Memory: 16GB RAN

Are you interested in submitting a PR to solve the problem?

• [x] Yes, I will!